Then, it finetunes the model by one/few-shot unseen image(s) in a self-supervised way to generate high-resolution (512 × 5 × 1024) results. In detail, it firstly trains a model in an extensive training set. To further improve the generalization ability of the unseen source images, a one/few-shot adversarial learning is applied. Furthermore, our proposed method can support a more flexible warping from multiple sources. Specifically, the source features are extracted by a denoising convolutional auto-encoder for characterizing the source identity well. To preserve the source information, such as texture, style, color, and face identity, we propose an Attentional Liquid Warping GAN with Attentional Liquid Warping Block (AttLWB) that propagates the source information in both image and feature spaces to the synthesized reference. It can not only model the joint location and rotation but also characterize the personalized body shape. In this paper, we propose to use a 3D body mesh recovery module to disentangle the pose and shape. However, they only express the position information with no abilities to characterize the personalized shape of the person and model the limb rotations. The existing task-specific methods mainly use 2D keypoints (pose) to estimate the human body structure. It means that the model, once being trained, can be used to handle all these tasks. We tackle human image synthesis, including human motion imitation, appearance transfer, and novel view synthesis, within a unified framework.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |